How To Make A Brain-On-A-Chip

Microfluidics comes of age across a wide swath of medical, industrial and military applications; after three decades the market finally takes off.


By Mark LaPedus
In October, Draper Laboratory and the University of South Florida (USF) disclosed an ambitious plan to develop a brain-on-a-chip.

The idea is to devise a “micro-environment’’ that mimics the human brain. Researchers hope to study neurodegenerative conditions such as Alzheimer’s disease, strokes and concussions. The eventual goal is to study the effects of drugs and vaccines on the brain.

Draper, a spinoff from the Massachusetts Institute of Technology (MIT), and USF are using embryonic cells from rats, but researchers plan to use human cells in the future. The brain-on-a-chip combines several technologies, including an emerging field called microfluidics.

Microfluidics deals with the control of fluids in devices. Tiny chip-like devices using microfluidics are used in many applications, such as cell sorting and detection, gene analysis, inkjet print heads, lab-on-a-chip units and point-of-care diagnostic tools. Meanwhile, lab-on-a-chip, and a related field, organ-on-a-chip (i.e. brain-on-a-chip), are systems that integrate various functions in a chip-like format. Some, but not all, lab-on-a-chip systems use microfluidics.

In these areas, OEMs and researchers use many of the same tools and processes borrowed from the semiconductor and MEMS industries. “The tools that are used to manufacture semiconductor devices can also be applied to make an organ-on-a-chip,” said Jeffrey Borenstein, distinguished member of the technical staff at Draper Labs. “The reason for that is you may need a lot of precision to structure these materials. The common tools are photolithography, etching, and other familiar processes in the semiconductor industry.”

Draper is not using leading-edge process technology, but it still faces some major challenges. “There is a need in the microfluidics field for process technologies that will enable things to be made at a higher volume and a lower cost,” said Borenstein, who is also the technical director for a separate project that is developing a human-on-a-chip. “There are some broader research goals in terms of understanding how organs work and understanding disease processes.”

Microfluidics emerges
Driven by the progress in the identification of genes and proteins, sales of microfluidic devices reached $1.3 billion in 2011, up 19% from 2010, according to Yole Développement. Yole also predicts that sales of fab-level microfluidics devices will average 23% annual growth through 2016, pushing the sector to almost $4 billion.

Microfluidics first emerged in the 1980s, but it has only recently begun to ramp up. The technology is being pursued by a plethora of device makers, OEMs and research entities. Affymetrix, Fluidigm, HP, IBM, Philips, Roche, Samsung, Siemens, Sony, STMicroelectronics are just a few of the names in the field.

In microfluidics, the products tend to be customized and the development cycles are long, but manufacturing costs are not an overriding issue. “A lot of people have set up fabrication capabilities for microfluidics that are relatively inexpensive,” said Draper’s Borenstein.

The tiny microfluidic devices themselves are generally comprised of complex pumps and external plumbing to transport a given fluid. The ability to pump the fluids at the micro-scale level is just one of the challenges. This has fueled the need for a new class of micro-pumps based on active and passive schemes.

“Microfluidics still suffers from the lack of a small, cheap and easy to integrate micro-pump,” said Alexander Govyadinov, an R&D engineer at Hewlett-Packard. “Generally, for a breakthrough to occur in microfluidic system development, essential microfluidic elements, pumps, valves, mixers and sensors, need to be integrated via low-cost fabrication technologies. There is also a lack of a killer application. A lot of progress happened, but even more development is required.”

There are other manufacturing challenges as well. “Most structures are 25 or 50 microns,” said Donald Johnson, chief executive of DJ DevCorp, a supplier of specialty dry film resist materials. “Now, there is a drive to get smaller dimensions. If you are looking at the channels, you want to get them down to a few microns. If you look at surface structures, you may be putting nano-structures on the surface.”

The initial, and many of the current, microfluidic devices are composed of silicon or glass and are made using etch and other processes taken from the semiconductor industry. “In many cases, the devices were manufactured using a single material, such as a microfluidic channel etched into a glass plate and sealed with a glass plate, yielding a monolithic microfluidic glass chip,” Johnson said during a recent presentation at an event sponsored by the Microelectronics Packaging and Test Engineering Council (MEPTEC).

The high cost involved in processing glass and silicon has prompted many vendors and research labs to use cheaper and transparent polymer materials like polydimethylsiloxane (PDMS). The polymer does not self-assemble, but rather the material is used to create the patterns or channels in a microfluidic device.

There are several ways to make a microfluidic device. In one common method, a vendor first makes a “master pattern” or “mold” of a microfluidic device. To make a mold, a resist, photomask and an absorber pattern are first applied on a substrate. Then, the “master mold” is patterned using laser ablation, micromilling or lithography.

Once the “master pattern” is created, the device is then replicated. There are several methods to replicate a device, such as casting, injection molding, thermoforming and hot embossing. A more advanced hot embossing technique is called nanoimprint lithography, which could shrink the channels down to the nano-scale. “You will probably see more embossing than imprint today,” Johnson said. “Nanoimprint is mainly in R&D at this point.”

Some are also using a technology called multi-layer soft lithography. In this process, a liquid polymer is poured over the mold. After a curing process, the polymer is peeled off the mold, leaving an imprint of the topography. Several layers of elastomer, which have different patterns, can be stacked and bonded. A microfluidic device usually has two elastomer layers for the flow and control functions.

New and emerging apps
The manufacturing processes and material choices depend on the application. For example, U.S.-based Micronics, which was recently acquired by Sony, last month launched a single-use, disposable card system that makes it possible to quickly determine a blood type.

Developed in part under funding from the U.S. Army, the company’s so-called ABORhCard is designed for a field-deployable test of potential blood donors in austere settings. “We were the first microfluidics company to move away from PDMS and glass to laminates. Final products are hybrid structures-laminates and injection mold,” said Karen Hedine, president of Micronics.

Another application is for a cell detection and sorting system. Cell sorting separates cells according to their properties. The microfluidics device in such a system can be done using micro-contact printing. “A well-known technique is done by contact printing, which can generate monolayers on the surface,” said Tohid Fatanat Didar, a visiting scholar within the Wyss Institute for Biologically Inspired Engineering at Harvard Medical School.

Microfluidics also plays a role in the emerging organ-on-a-chip field. Last year, the National Institutes of Health (NIH), the Defense Advanced Research Projects Agency (DARPA), and the U.S. Food and Drug Administration (FDA) announced plans to develop a chip technology that could screen drugs and vaccines more rapidly and efficiently than current methods. The chip is based on specific cell types that reflect human biology. NIH will commit up to $70 million and DARPA will commit a comparable amount.

DARPA recently awarded a contract to Draper through MIT. Draper is working on various projects such as a human-on-a-chip, brain-on-a-chip, liver-on-a-chip, and a kidney-on-a-chip.

The brain-on-a-chip project itself combines cellular neuroscience, tissue engineering and microfluidics. The chip aggregates cultured neurons, astrocytes, microglia and brain endothelial cells from rats on two micro-fabricated layers. A microfluidic pump was used to circulate nutrients or therapeutics across the vascular channels simulating blood flow. “What we’re trying to do is take the most important features of an organ and scale them down to the micro-scale. Then, you put them on some kind of a platform, where you can evaluate them,” said Draper’s Borenstein.

The fabrication process is straightforward. “You start with a silicon wafer process. And then you transfer that into a polymer using some kind of a molding or embossing technique,” he said. “The bigger challenge is to transition this from a silicon to a non-silicon base. We do not use the latest and greatest photolithography. In fact, most of the photolithographic work that’s done in our lab is done with structures that may have a minimum dimension of 10 microns.”

Still, Draper and other entities will need new manufacturing breakthroughs. “Hot embossing is a good example. Hot embossing is a great process, but it’s a little bit slow,” he said. “The material that has carried the microfluidic field for many years is called PDMS. It’s very inexpensive and easy to process, but it’s not particularly stable chemically.”

Researchers are looking at new and advanced bio-degradable materials for future work. And, of course, the field of organ-on-a-chip technology is still in its infancy. “It’s still in the very early stages,” said Anil Achyuta, principal investigator for the brain-on-a-chip project at Draper. “We have the potential to revolutionize how scientists study the effects of drugs, vaccines, and specialized therapies like stem cells on the brain.”

Leave a Reply